PaperHub
影响力指数
85.77/100
前 0.9%
全站排名 #556
发表论文20 篇
平均评分6.0
年均产出10.0 篇/年

Binhang Yuan

Assistant Professor@Hong Kong University of Science and Technology·中国香港·OpenReview
研究方向

Large language model · Distributed machine learning system · Distributed stochastic gradient descent · Database system

7.3
17

FSA: An Alternative Efficient Implementation of Native Sparse Attention Kernel

ICLR 2026Poster
通讯
6.8
21

Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes

ICLR 2026Poster
6.0
21

Unlocking Full Efficiency of Token Filtering in Large Language Model Training

ICLR 2026Poster
6.0
24

UltraLLaDA: Scaling the Context Length to 128K for Diffusion Large Language Models

ICLR 2026Poster
通讯
5.5
15

Cascadia: An Efficient Cascade Serving System for Large Language Models

ICLR 2026Poster
通讯
4.5
24

Local Success Does Not Compose: Benchmarking Large Language Models for Compositional Formal Verification

ICLR 2026Poster
通讯
4.5
29

TAH-QUANT: Effective Activation Quantization in Pipeline Parallelism over Slow Network

ICLR 2026Rejected
通讯
4.0
29

TQA-Bench: Evaluating LLMs for Multi-Table Question Answering

ICLR 2026Withdrawn
4.0
18

VeriEquivBench: An Equivalence Score for Ground-Truth-Free Evaluation of Formally Verifiable Code

ICLR 2026Poster
4.0
12

Parallax: Efficient LLM Inference Service over Decentralized Environment

ICLR 2026Rejected
通讯
8.2
26

MeCeFO: Enhancing LLM Training Robustness via Fault-Tolerant Optimization

NeurIPS 2025Poster
8.2
17

AREAL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning

NeurIPS 2025Poster
7.5
21

DeFT: Decoding with Flash Tree-attention for Efficient Tree-structured LLM Inference

ICLR 2025Spotlight
6.8
29

Multi-step Visual Reasoning with Visual Tokens Scaling and Verification

NeurIPS 2025Poster
6.8
24

Efficient Pre-Training of LLMs via Topology-Aware Communication Alignment on More Than 9600 GPUs

NeurIPS 2025Poster
6.4
32

Hallucination at a Glance: Controlled Visual Edits and Fine-Grained Multimodal Learning

NeurIPS 2025Poster
通讯
6.0
28

HexGen-2: Disaggregated Generative Inference of LLMs in Heterogeneous Environment

ICLR 2025Poster
三作
5.8
40

Locret: Enhancing Eviction in Long-Context LLM Inference with Trained Retaining Heads

ICLR 2025Rejected
二作
5.5
9

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs

ICML 2025Poster
5.5
45

Multi-Agent Collaborative Data Selection for Efficient Language Model Pretraining

ICLR 2025Rejected
合作者 (20)